A Review of Criticality Accidents A Review of Criticality Accidents

time of the accident. However, the investigation was
not able to determine who made the transfer or when it
had taken place.
As part of the investigation, both experiments and
calculations were performed to estimate the conditions
for criticality in vessel 18. The results determined that
30 l of solution containing 825 g of plutonium
(27.5 g/ l) would be required for criticality. These
values agree closely with the best estimate of the
contents of vessel 18 at the time of the accident, 31 l
of solution with 848 ± 45 g of plutonium (27.4 g/ l).
One contributing cause of the accident was the
unrecorded transfer of 5 l of solution from vessel 1 to
vessel 18.
The accidental excursion resulted in approximately
2 × 10 17 fissions. This estimate was based on a
temperature increase of 60°C in 31 l of solution. The
2. Mayak Production Association, 21 April 1957
This accident occurred in a large industrial building
housing various operations with highly enriched
uranium. Operations were being conducted under the 6
hour shift, 4 shifts per day mode prevalent at Mayak.
Rooms typically contained several gloveboxes separated
from each other by about two meters and
interconnected by various liquid transfer and vacuum
lines. The accident took place in a filtrate receiving
vessel that was part of batch mode, liquid waste
processing and recovery operations.
A layout of the glovebox and its equipment is
shown in Figure 7. This was a typical one workstation
deep by two workstations wide glovebox. The normal
process flow was as follows: the main feed material,
impure uranyl nitrate, was generated in upstream
U(90) metal purification operations. This, along with
oxalic acid, was introduced into the precipitation
vessel, which was equipped with a stirrer and an
external steam/water heating jacket. A batch would
typically contain a few hundred grams of uranium feed
in about 10 l of liquid; concentration was usually in
the 30 to 100 g U/ l range. The stirrer operated
continuously during the process to prevent the accumulation
of oxalate precipitate on the vessel bottom.
Precipitation of the uranyl oxalate trihydrate proceeded
according to the following reaction:
60°C temperature rise was based on the coarse
observation that the solution following the accident
was at or near the boiling temperature. The accident
caused no physical damage to any equipment. The
operator positioned in the cell received an estimated
dose of 100 rad. The operator near vessel 18 received
an estimated dose of 1,000 rad. He suffered severe
radiation sickness and amputation of both legs. He died
35 years after the accident.
Procedures in place before the accident were
unambiguous in specifying that vessels 2, 4, and 6
were to never contain solution. The presence of
solution in vessels 2 and 4 at the beginning of the shift
prior to the accident illustrates that procedures were
being violated. The entries in Table 1 also shows that
the mass limit of 500 g per vessel was being violated.
Uranium precipitate, U(90), buildup in a filtrate receiving vessel; excursion history unknown; one fatality, five
other significant exposures.
( ) + + → • ↓
UO2 NO3 H2C2O4 3H2OUO2C2O4 3H2O
+2HNO
2
3
The oxalate precipitate slurry was then vacuum
transferred to a holding tank from which it was drained
into a filter vessel. The precipitate containing the
uranium was collected on the filter fabric, and the
filtrate was pulled through by vacuum and collected in
a filtrate receiving vessel, where the accident took
place. This vessel was a horizontal cylinder 450 mm in
diameter by 650 mm in length and had a volume of
approximately 100 l. As indicated in the figure, the
filtrate was removed through a dip tube and transferred
to an adjacent glovebox.
A two tier hierarchy of procedures and requirements
was in place at the time. Upper level documents
described operations covering large work areas in
general terms, while criticality guidance was contained
in operating instructions and data sheets posted at each
glovebox. Specifics associated with each batch, such as
the fissile mass, time, temperature, and responsible
operators, were recorded on the data sheets that were
retained for one month. Important entries from the data
sheets were transcribed to the main shift logs that were
retained for one year.
Operational and fissile mass throughput considerations
dictated the design and layout of glovebox
equipment. Thus, major pieces of equipment were not
necessarily of favorable geometry. Limitation of the
fissile mass per batch was the primary criticality
control throughout the glovebox. The procedure called
9